Diabetes Medications Improving, But Not Blood Sugar Levels

A new study shows that choice of diabetes medications is changing in the U.S. The new drugs are supposedly better, but average blood sugar levels appear to be going up. How do you explain this? Are drugs not the answer?

The study by Kasia J. Lipska, MD, MHS, and colleagues at Yale University, Mayo Clinic, and the University of California, San Francisco, was published in Diabetes Care in September. Researchers reviewed records of over 1.6 million privately insured and Medicare Advantage patients from 2006–2013. They recorded what drugs patients were taking, their HbA1c levels (a measure of average glucose for the last 2–3 months), and the frequency of severe hypoglycemia (HIGH-poe-gly-SEEM-e-uh, or low blood sugar) episodes.


The insurance records showed that doctors are prescribing more DPP-4 inhibitors, insulin, and metformin. They are using less of the sulfonylureas such as glimepiride, glipizide, and glyburide. Use of another drug category, thiazolidinediones, or TZDs has practically stopped. Only 5.6% of people in the study were receiving these drugs by 2013.

There have been good reasons for cutting back on TZDs and sulfonylureas. TZDs were found in several studies to cause heart failure in vulnerable patients. Sulfonylureas cause the pancreatic beta cells to release insulin whether or not blood sugar levels are rising and so put people at risk for hypos.

The older drugs also had advantages. TZDs reduce insulin resistance and blood pressure. They raise levels of HDL (“good”) cholesterol. Sulfonylureas are cheap and convenient to take, so people are more likely to use them.

People are using more insulin, DPP-4 inhibitors, and metformin now. Are those drugs better, or are they just in fashion?

Their reviews sound good. Insulin lowers blood sugar by taking sugar into cells. Metformin helps the liver hold onto extra glucose instead of spilling it and makes the body’s cells more sensitive to insulin. DPP-4 inhibitors raise levels of hormones called incretins. Incretins improve insulin production, slow stomach emptying, and prevent the liver from releasing glucose. Seems these drugs should be making things better.

Are they helping?
When it comes to glucose control, though, the changed prescriptions haven’t made a difference. If anything, average glucose levels measured by HbA1c are going up. “The proportion of patients with HbA1c <7% ["good control"] declined from 56.4 to 54.2%," according to the study. "The proportion with HbA1c ≥9% [poor control] increased from 9.9 to 12.2%."

Control varied by age. Roughly 23.3% of young patients had poor control, compared to about 6.3% of older patients. The study found that rates of hypoglycemia were unchanged at about 1.3 events per 100 person-years. That means that if you followed 50 patients for two years, you would expect to see between 1 and 2 cases of severe hypoglycemia in that time.

Why aren’t better drugs leading to lower glucose levels? It could be that people rely on their medications instead of on their self-management. They get a new drug and think, “Great. Now I can eat whatever I want.” Or maybe, “I don’t need to exercise; these pills have got me covered.”

Or it could be that insulin and metformin are not actually better glucose-lowering drugs than sulfonylureas and TZDs, if those are used correctly. It could even be that, because some of the popular drugs are more expensive, economic stress cuts into some people’s healthy living. TZDs and sulfonylureas are available in generics, and some cost just a few cents a pill. Metformin is also cheap, but DPP-4 inhibitors and synthetic insulin analogs can cost hundreds of dollars a month.

The good news is that rates of complications like heart disease, stroke, and lower-limb amputation have not changed much recently, and they are much lower than they were about 20 years ago. Glucose control isn’t everything, though it is the best indicator we have of diabetes status.

Should data like this new study affect your and your doctor’s choice of medications? Should you stop taking some or all of your medications, if we don’t know which ones are best? I would say not to pay too much attention to large-scale studies and pay more attention to how individual drugs affect you. Not just your glucose numbers, but your overall health and how you feel. Consider side effects, cost, and all your lab numbers, not just glucose.

The main point is that no drug is going to solve diabetes for you, but you can make it much better with good self-management.